[This article originally appeared in The Ergonomics Report™ on July 1, 2002, over 10 years ago. Stay up-to-date with the evolving world of ergonomics by subscribing to The Ergonomics Report™ today.]
Designers are often frustrated by humans. It sounds funny, because nearly all designs are destined to be used — or misused — by people, but it’s true. Designing something to accommodate for the wide variety of human sizes and shapes can be so frustrating that many designers simply neglect to do it. Some neglect to accommodate intentionally, others try, yet make mistakes in the process, and still others neglect it out of pure ignorance.
Intentionally choosing not to design a product for each and every person in the world is common, necessary, and acceptable. Informed businesses, policy makers, designers and ergonomists do it and recommend it every day. Trying to accommodate a variety of people, yet failing, can be due to either misapplication of available knowledge, lack of access to available knowledge, or gaps in that knowledge base. Although it occurs more often than it should, there is no excuse for failing to accommodate out of pure ignorance.
Anthropometry is the knowledge base we apply when trying to understand and accommodate for the physical differences among people. Dr. Bruce Bradtmiller, President of Anthrotech, Inc., gives the simple example of how one might design a doorway using such data. If we can fit the expected height and width of people, we have the two dimensions we need to proceed. Sounds simple, right?
Actually, even such simple problems are made complicated by real life. For example, we would want to consider allowances for clothing, shoes, hats, safety and emergency clearances, room for items that may be carried through the door, etc.
For one or two dimension design problems like the door example, we typically turn to simple “percentile” data sets. For example, if we decide to fit the height of 95 percent of all Americans through our doorway, we would select 95th percentile American height, plus allowances, as our design dimension.
What if we’re designing something that has more than one or two important dimensions — an office chair, for example? We’d want to fit the length of peoples’ lower legs, including shoes, to determine seat height; upper leg lengths to determine seat pan depth; buttocks and hip widths to determine seat pan width; and so on. Unfortunately, we can no longer rely on the percentile data we used to design our doorway, because we now have multiple dimensions, or multiple variables to consider, not just height and width. If we hope to achieve a good fit, we need to turn to multivariate anthropometry methods which rely on statistical approaches to account for variability among the multiple dimensions. In other words, multivariate approaches help us account for the fact that people who are, for instance, 50th percentile, or “average” height, do not necessarily have 50th percentile buttocks widths. If we assumed they did, we’d likely end up with a design that didn’t fit anyone well. (Chair manufacturers have discovered that it is indeed difficult, even impossible, to provide a good fit for everyone with just one chair model. Many have made the informed business decision to offer big, medium, and small versions to better accommodate more people.)
Researcher Kathleen Robinette, of the U.S. Air Force Research Laboratory, explains that multivariate anthropometry methods provide a methodology through which we select the most important dimensions for a specific product based on the intended user population, the tasks those users will be expected to perform, and environment of use. “Multivariate methods allow us to consider two or more dimensions simultaneously,” adds John Roebuck, President of Roebuck Research and Consulting.
It’s good to know that there are data and techniques available to help designers better fit the people they design for, but finding the data is not always easy. Sometimes, it just doesn’t exist. And sometimes the dimension thought to be appropriate turn out to be wrong.
For example, “Women’s hip circumference is usually largest down on the thighs,” says Robinette. “Women consider that legs, not hips, but that’s where most are widest. For men, it’s higher up, actually on the torso, not the thighs.”
Americans Are Getting Fatter
“The historic data indicate we were getting taller until about the mid-1970s,” says Robinette. “Now we don’t seem to be getting taller, but we do seem to be getting fatter.”
According to some estimates, about 26% of American adults are obese. That’s about 54 million adults that are roughly 30 pounds or more over what is considered a healthy weight.
There’s a similar trend among American kids. The percentage of significantly overweight black and hispanic children has more than doubled in 12 years. There was a 50% increase among white children. As of 1998, among kids 2 to 12 years old, almost 22% of blacks, 22% of hispanics, and 12% of whites were overweight.
According to Bradtmiller, deciding who to accommodate in a design is a business or policy decision. Recent articles in the news demonstrate the growing consequences weight-based design decisions may bring.
For instance, Southwest Airlines made its business decision clear when it announced a policy requiring the purchase of two tickets if the customer tends to spill over into their neighbor’s seat. In fact, most economy class air travelers will report discomfort in airplane seating, either due to their own size, the size of neighboring passengers, or simply due to the seat design itself. It seems that nobody feels a good fit in a typical airplane seat. If the obesity trends continue, we can speculate it won’t get any better unless airlines pay more attention to — and spend more money on — seat design. And if they provide more space to passengers, they will be able to fit and sell less seats per flight, increasing the economic impact. So, especially with the airline businesses experiencing an economic downturn, don’t expect the situation to get any better soon — that is, unless you are willing to pay for the increased costs, which the airlines are currently banking you won’t.
Another example, reported in Ergonomics Today™ (November 2001), described the events surrounding a woman’s death at Knott’s Berry Farm, an amusement park in California, USA. According to the article, a 40-year-old woman weighing nearly 300 pounds fell to her death from the “Perilous Plunge” water ride. The manufacturer believes the woman fell from the ride because she was not properly belted, and should not have gotten on the ride if she could not be safely restrained. The family’s attorney claims that abrasions on her thighs indicate that she was belted in before she fell. The coroner reported that her abdomen was eight inches larger than the 50-inch seat belt, and investigators speculated that if belted, she may have placed the belt below her abdomen.
And then, a June 2002 article in Ergonomics Today™ summarized a study published in Accident Analysis and Prevention that found heavier people are more likely to be killed or seriously injured in car accidents than lighter people. Specifically, people weighing between 100 and 119 kg (220-262 lbs.) are almost two-and-a-half times as likely to die in a crash as people weighing less than 60 kg (132 lbs). The study concluded that it’s not just total weight, but obesity that increases the risk.
Theories abound to explain this finding. Some suggest it shows that crash tests, and therefore car interior designs, do not accurately reflect the people who may be in the car, because crash tests typically use dummies that represent a “standard-sized” male weighing about 78 kg (172 lbs). Others suggest that obese people, with health problems such as high blood pressure or diabetes, could be finding it tougher to recover from injury. Still others think the link might be due to poor seat belt restraint during a crash. One researcher suggested that a large amount of fat tissue between the restraint system and the bony thorax introduces “slack” into the restraint system, decreasing its performance. (Interestingly, among five health risk measures recently analyzed by Behavioral Risk Factor Surveillance System, obesity was up in all 47 states surveyed, and seat-belt use was up in 39 of the 47 states.)
The Bottom Line
Regardless of the reasons, it seems that designers cannot ignore the fact that at least some American populations are getting more obese. Obesity trends undoubtedly have health implications, but what implications might they have for designers? Do we change our physical world — chairs, desks, tools, etc. — to fit changing human dimensions, or do we change people to fit the physical world? In other words, do we put obese people on weight control regimes, or do we recognize these trends and get busy designing for them?
The answer is, of course, it depends. It really comes down to a business or policy decision. Public health officials may argue that society should do what it takes to get people to stay fit and trim. But, in civil societies there are few, if any, options available to force people to do something against their will. A civil society cannot force people to lose weight. A decent employer or government cannot force people to exercise and eat less. The only way to successfully accomplish goals that rely on human behavior is to influence individuals to take it upon themselves to, in this case, lose weight and keep that weight off. In the mean time, designers cannot sit back and wait for people to lose weight. It is his or her job to adapt systems to humans, not vice-versa. Persuading people to be physically fit is laudable, but it is not an acceptable approach to ergonomics on its own merit.
Design discrimination happens every day, will continue to happen, and should continue to happen. Ergonomists, engineers, architects and other designers routinely make design decisions that discriminate. If it’s an informed design, there’s nothing inherently wrong with it, and it’s usually unavoidable.
Designers and organizations who neglect ergonomics principles — and unfortunately there are many — may unintentionally discriminate against a large number of users because they don’t adequately understand or consider human variations in their designs. But even when ergonomics is thoroughly considered, it’s typical to design only for a reasonable segment of the expected user population. How to reduce risk after a design is complete, such as, user and use restrictions, training, warnings, etc., is another subject directly impacted by ergonomics, but beyond the scope of this article.
If American populations are growing in girth, can your organization afford to ignore the trend? It may be time to closely evaluate business and policy decisions that are impacted by obesity.
Thanks to Bruce Bradtmiller, Ph.D., President of Anthrotech, Inc. (www.anthrotech.com), Kathleen Robinette, Director of the CAESAR 3D Anthropometry Project and Principle Research Anthropologist, Air Force Research Laboratory, John Roebuck, President of Roebuck Research and Consulting, and Rachel Michael, an ergonomist formerly with Ergoweb Inc., now with Marsh, for assisting with this article.